By Orren Peled
VP of Research & Development—Global Risk Solutions
Harland Financial Solutions Worldwide
This e-mail address is being protected from spam bots, you need JavaScript enabled to view it • 800-815-5592 • www.harlandfinancialsolutions.com

 

 

 

Over the past few years, Basel II has spurred financial institutions to transform their regulatory capital reporting capabilities. In the rush to meet regulatory deadlines, financial institutions had to retrofit the new regulatory calculation and reporting systems onto existing legacy infrastructures. Stepping back to look at what they have accomplished, some financial institutions are witnessing mixed results.

Ironically, while larger banks contemplate their existing systems, smaller banks and banks in emerging economies can benefit from the collective experience, and move ahead faster to implement new credit origination processes that surpass those commonly found in the largest institutions.

The basic problem is that credit origination processes that feed the regulatory capital requirements have remained a patchwork of systems and disparate data silos. Hence, they are reducing business process efficiencies and increasing the IT cost of ownership. These processes cover a wide range of systems, from front office sales to back office booking of credit transactions on the host system.

Ask a group of bankers to identify the number of different systems they use to manage credit risk and to originate commercial loans, and the list quickly grows.  Here are some probable responses:

•    Core processing system
•    Customer relationship management (CRM) system
•    Financial statements analysis system
•    Covenant tracking system
•    Underwriting and credit write-up system
•    Credit risk rating system
•    Credit pricing system
•    Credit approval workflow system
•    Limits management system
•    Tickler and exception reporting system
•    Document management system
•    Loan document preparation system
•    Line of business portfolio reporting system

 
Typically, some of these systems are sourced from external vendors offering point solutions. In-house software teams will develop other systems in addition to semi-manual solutions to fill the remaining gaps. The result: financial institutions are managing their most important asset—customer credit risk information—in many disparate systems and data silos.

Furthermore, if we take a closer look at the commercial credit origination process, we see that it spans a diverse set of customer profiles—from small businesses, through mid-market enterprises to large corporate entities. In addition, customers belong to many different industry types. These differing customer segments demand specific data acquisition, scoring and decisioning strategies. This diversity has often led to the development of multiple systems that support different customer profiles and product lines.

The strain that such a heterogeneous environment imposes on end users is clear. Users spend more time searching for data and rekeying information, mistakes are more common, workflows slow down and frequent training is required to keep users adept across all systems. As a result, time better spent analyzing and evaluating the risk/return of the originated credit is lost as frustrated users chew on the smorgasbord of systems confronting them.

For the IT departments that support these systems, the challenges are no less demanding. Installing, maintaining, upgrading and backing up the various systems only increases the total cost of ownership (TCO). Systems developed in-house often rely on a few key individuals who understand the complex business logic under the hood. As the business requirements evolve, and key personnel dissolve, it becomes increasingly difficult to implement changes in these systems.  Indeed, financial institutions become reluctant to touch the delicate integration components that bind the various systems together. Modifying any one system is often perceived as being risky and to be avoided or delayed if possible.

To improve matters, financial institutions will need to rethink and then re-engineer their credit origination systems and processes. This may sound unattainable, but it is not. In most cases, we are talking about a gradual transition that takes into consideration the budgetary and resource constraints faced by all. Start by evaluating your current situation, and then set about developing a roadmap to attain longer-term objectives. The following sections offer some guidance and useful discussion points:

 

• Identify the stakeholders: Relationship managers and lenders from the main lines of business, credit analysts, portfolio risk managers, debt recovery staff, legal and compliance staff, finance directors and, of course, IT staff. Managing these diverse participants can be quite a challenge; however, their input will help convince you how integrated and overlapping their needs are.

• Together with the stakeholders, define in advance the key metrics that will be used to measure exposure levels, risk ratings, profitability, regulatory and economic capital usage and performance management.

 

• It will soon become evident that only a unified data architecture can support the required metrics and underlying processes. Let's look at two examples to illustrate the point:

 

1. In order to price a credit, we need access to the following metrics: exposure levels, rates and fees, risk ratings (PD, LGD), operational expenses and capital usage. Each one of the preceding metrics is derived from many supporting data elements. For real-time pricing, we need real-time access to all of the underlying data elements. Only then will commercial lenders be able to run pricing scenarios and optimize deal structures.

 

2. A customer submits a fresh set of financial statements that breach a financial covenant set by the bank. This automatically triggers an exception and causes an automatic downgrade in the risk rating of the customer. In addition, a workflow to contact the customer initiates. This is followed by a manual review of the customer and a re-rating. In many institutions, this perfectly common incident may require a user to access half a dozen different systems to deal with the issue. Neglecting the incident, because the processes are not in-sync, may have undesirable consequences.

 

• A superior credit risk management system must have a long memory. Historically accumulated data is vital for risk modeling and back testing. The relevant model input parameters need to be captured and stored over several years. This is because the experts who design and test models will usually want the data underpinning the models to be at the most granular level. The quality and completeness of the data are also very important. A well-designed data architecture will facilitate more predictive scorecards and risk models that will pay back for themselves many times over.

 

• Avoid a piecemeal approach. Disparate systems will never deliver an efficient and streamlined solution. Look for infrastructures that provide an enterprise-wide solution by eliminating stand-alone point applications. Compare existing stand-alone applications with the functionality offered by best-of-breed credit origination suites. If the solution offered by the suite matches or surpasses the stand-alone application, the choice is obvious.

 

• A comprehensive solution may still need to be implemented in stages. Therefore, the solution should be modular, allowing the financial institution to replace key components in serial order. This will allow time for integration and data migration work to be scheduled more efficiently.

 

• Credit origination systems require several important integration points with external systems. These will usually include the following:
- Host loan processing system(s) and central customer information file
- Risk data warehouse to support reporting, model development and validation
- Links to external providers such as credit bureaus, rating agencies, title and lien data providers
- Single sign-on mechanism to administer user access rights

 

• Reuse existing infrastructures when it makes sense to do so. For example, in some cases, it may prove advantageous to integrate generic enterprise-wide workflow management tools and document management systems with the credit origination system. In other cases, it may be more efficient to use the built-in, dedicated features that exist in the credit origination suite.

 

• Select systems that employ loosely coupled Web services or a service-oriented architecture (SOA) to integrate with other systems. This will shorten integration cycles and improve robustness.

 

• To cope with diversity, credit origination systems must be flexible and meet a broad set of functional requirements. Credit risk best practices are constantly changing; therefore, the origination system will need to be adaptable as well. Deploy systems designed to use meta-data definitions for configuration. Areas that should be entirely configurable include:
- Financial analysis models and worksheets
- Credit policy checklists
- Credit products and collateral items
- Risk-rating scorecards and models
- Credit analysis and approval workflows
- Credit write-up formats
- Analysis and reporting outputs

Challenge vendors to prove how close their systems come to meeting all your financial institution’s requirements without resorting to hard-coded customizations.

 

• Find out whether checklists, analyses and workflows can be tailored to specific deal types. Does the workflow support branching, load balancing, decision points and time-outs?

 

• Are reusable templates available to pre-populate narrative and worksheet areas?

 

• Does the system automatically and fully archive all historical credit applications?

 

• Check out the included configuration tools. How flexible is the configuration and how many tables are modifiable? Can the financial institution do this independently?

 

• Customizations may be required in some instances. Verify that there are well-developed APIs and that any customizations will survive future version upgrades without requiring any additional work.

 

• Can the financial institution safeguard its proprietary models, templates and processes? A flexible and configurable system based on meta-data definitions will allow a financial institution to configure the systems and enhance its processes based on its accumulated internal experience and expertise with negligible risk of knowledge transferring to competitors.

 

• Ensure that granular functional level security exists throughout the systems. Credit origination is a very collaborative process with participants having clearly defined roles and authority levels. All vital transactions should generate an audit trail. Credit approvals (and declines) should lock down the pertinent data. These mechanisms serve to lower operational risks.

 

• Hand off the business logic administration to the business users. Enabling certified business administrators to configure and manage the systems reduces the burden on IT and allows the business more freedom.

 

• Browser-based front ends are now the norm, but not all are equal. Look for systems that make efficient use of the user interfaces, are intuitive for users and promote a fast learning curve. User-defined settings and configurable context-sensitive help should be standard features.

 

• The solution should also support peripheral support teams within the bank, such as the loan review department and the debt recovery group. These functions need access to exactly the same information that exists in the credit origination process. They should not be using separate systems.

• Choose vendors with domain expertise and proven experience. Credit origination and risk management are complex environments that will punish newcomers who underestimate the full scope and complexity of the project.

• IT departments have a vital role in managing the transition. Bringing together the stakeholders and solution providers should generate a lively debate. The outcome of these brainstorming sessions will help the financial institution determine the preferred course of action.

Credit origination technology has matured. Systems are finally bringing together all the main processes under one roof and delivering the benefits that lenders have been talking about for a long time. Financial institutions can now start executing strategies that will lead to noticeable improvements in productivity and provide the basis for sounder enterprise risk management.

About the Author

Orren Peled is Vice President of Research and Development with the Global Risk Solutions unit of Harland Financial Solutions Worldwide. His expertise is in the area of enterprise decision support systems with a primary focus in the banking industry. He has extensive experience as a System Analyst and Consultant in the banking industry. For more information about how to streamline your credit origination process, visit www.creditquest.com or send an e-mail to This e-mail address is being protected from spam bots, you need JavaScript enabled to view it

 

Name

Email

Title

Comment smaller | bigger

Subscribe via email (Registered users only)

I have read and agree to the Terms of Usage.

Add Comment